Continuity of daily temperature time series in the transition from conventional to automated stations for the Colombian coffee network

The transition from conventional weather stations (CWSs) to automated (AWSs) of the Colombian coffee network has required testing their performance and adjusting temperature measurements ensure continuity historical CWS series. In this study, mean (Tmean), minimum (Tmin), maximum (Tmax) AWS operating in parallel at 36 locations between 2014 2019 were compared, biases daily differences (CWS − AWS), agreement index (d), percentage data within allowed range (PR05) calculated. most consistent method for calculating Tmean Tmax was selected use on data. With standard normal homogeneity test with metadata, we found that series not homogeneous, instrument failures sensor changes being main causes lack homogeneity. statistical analyses indicated need be adjusted continuous To correct bias, two approaches applied: quantile mapping additive constant. results suggest adjustments improve average bias all but do necessarily bring ±0.5°C. Tmin Tmean, 12 AWSs can give CWS, rest variables, are independent CWSs.